Article
Identification of Long Non‐Coding RNAs Deregulated in Multiple Myeloma Cells Resistant to Proteasome Inhibitors Ehsan Malek 1, Byung‐Gyu Kim 2 and James J. Driscoll 3,4,* Division of Hematology and Oncology, University Hospitals, Case Medical Center, Seidman Cancer Center, Cleveland, OH 44106, USA;
[email protected] 2 Department of Pediatrics, Case Comprehensive Cancer Center, Case Western Reserve University, Cleveland, OH 44106, USA;
[email protected] 3 Division of Hematology and Oncology, University of Cincinnati College of Medicine, Cincinnati, OH 45267, USA 4 The Vontz Center for Molecular Studies, University of Cincinnati College of Medicine, Cincinnati, OH 45267, USA * Correspondence:
[email protected]; Tel: 513‐558‐2186 1
Academic Editor: Muller Fabbri Received: 5 July 2016; Accepted: 27 September 2016; Published: 6 October 2016
Abstract: While the clinical benefit of proteasome inhibitors (PIs) for multiple myeloma (MM) treatment remains unchallenged, dose‐limiting toxicities and the inevitable emergence of drug resistance limit their long‐term utility. Disease eradication is compromised by drug resistance that is either present de novo or therapy‐induced, which accounts for the majority of tumor relapses and MM‐related deaths. Non‐coding RNAs (ncRNAs) are a broad class of RNA molecules, including long non‐coding RNAs (lncRNAs), that do not encode proteins but play a major role in regulating the fundamental cellular processes that control cancer initiation, metastasis, and therapeutic resistance. While lncRNAs have recently attracted significant attention as therapeutic targets to potentially improve cancer treatment, identification of lncRNAs that are deregulated in cells resistant to PIs has not been previously addressed. We have modeled drug resistance by generating three MM cell lines with acquired resistance to either bortezomib, carfilzomib, or ixazomib. Genome‐wide profiling identified lncRNAs that were significantly deregulated in all three PI‐ resistant cell lines relative to the drug‐sensitive parental cell line. Strikingly, certain lncRNAs deregulated in the three PI‐resistant cell lines were also deregulated in MM plasma cells isolated from newly diagnosed patients compared to healthy plasma cells. Taken together, these preliminary studies strongly suggest that lncRNAs represent potential therapeutic targets to prevent or overcome drug resistance. More investigations are ongoing to expand these initial studies in a greater number of MM patients to better define lncRNAs signatures that contribute to PI resistance in MM. Keywords: long non‐coding RNA; multiple myeloma; myelomagenesis; proteasome; drug resistance
1. Introduction Multiple myeloma (MM) is characterized by the clonal proliferation of malignant plasma cells (PCs) within the bone marrow (BM) microenvironment [1,2]. MM is the second most common hematologic malignancy, with an incidence of 24,000 cases per year in the USA, and accounts for 2% of deaths from all cancers and ≈20% of deaths from hematologic cancers [3]. In Western countries, in the near future both the incidence and prevalence of this disease will increase as the result of an aging Genes 2016, 7, 84; doi: 10.3390/genes7100084
www.mdpi.com/journal/genes
Genes 2016, 7, 84
2 of 13
population and increased survival of those with the disease. As the number of patients surviving with MM increases, concern for the development of therapeutic resistance rises. Although the introduction of immunomodulatory drugs (IMiDs) and proteasome inhibitors (PIs) has improved the outlook, MM remains incurable. Induction therapy followed by autologous BM transplant forms the backbone of treatment for young and transplant‐eligible patients. However, many patients do not respond to standard therapy and those that do respond inevitably develop resistance [4]. Eukaryotic cells maintain a healthy proteome through integration of the ubiquitin (Ub) + proteasome system (UPS) and aggresome + autophagy pathway [5]. The UPS plays a pivotal role in maintaining proteostasis through the selective elimination of misfolded, damaged, and short‐lived proteins [6–8]. Molecules that inhibit the proteasome and disrupt proteostasis are selectively cytotoxic to cancer cells and have been exploited for therapeutic gain. Functional blockade of the UPS represents a remarkable bench‐to‐bedside success that has catapulted the proteasome into a position of prominence in cancer biology [9–12]. PIs such as bortezomib (Velcade™, Takeda Oncology, Cambridge, MA,) and carfilzomib (Kyprolis®, Amgen, Thousand Oaks, CA) have improved the quality of life (QOL) and overall survival (OS) of MM patients. Bortezomib is a selective drug that inhibits the proteasome to exploit its pivotal cellular function and promotes tumor cell death. Ixazomib (Ninlaro®, Millennium, Takeda Oncology, Cambridge, MA) is a boron‐based, orally available PI, and was recently Food and Drug Administration (FDA) approved [13–16]. PIs also induce aggresome formation and autophagy as compensatory protein clearance mechanisms and this can lead to the emergence of drug resistance and disease relapse, leading to treatment failure and fatal outcomes. While the therapeutic benefit of targeting the proteasome remains unchallenged, more precise modalities that do not induce the aggresome + autophagosome pathway are needed. Only one‐fifth of transcription leads to protein‐coding genes across the human genome, indicating that there are four times as many non‐coding RNAs (ncRNAs) compared to coding RNAs. Long non‐coding RNAs (lncRNAs) function as master regulators of the human genome and control the majority of key intracellular signaling pathways and processes including development, proliferation, differentiation, and apoptosis [17–19]. Accordingly, alterations in ncRNA expression are seen in tissues and contribute to cancers, autoimmune and genetic disorders, and infectious processes [20–24]. Specifically, deregulation of ncRNA levels is linked to tumor initiation, metastasis, and drug resistance. Thus, ncRNAs have rapidly attracted attention as potential diagnostic and therapeutic targets. lncRNAs are transcripts longer than 200 nucleotides. This arbitrary limit differentiates this group from smaller regulatory RNA such as microRNAs (miRNAs), small nucleolar RNAs (snoRNAs), Piwi‐interacting RNAs (piRNAs), and short interfering RNAs (siRNAs). Certain ncRNAs may represent new therapeutic targets to overcome drug resistance in cancer [25– 27]. MM is characterized by genetic heterogeneity in terms of changes in gene expression compared to normal BM plasma cells [28–30]. These expression changes can be driven directly or indirectly by changes in signaling, e.g., due to altered external stimuli mediated by a changing microenvironment such as by miRNAs [31]. miRNAs are non‐protein‐coding RNAs that function as regulators of mRNA stability and translation. miRNAs act post‐transcriptionally to repress the expression of their target genes, while an upregulation of gene expression has been found under specific conditions, e.g., with specific transcripts, in distinct cell types [32]. A single miRNA is typically involved in the regulation of several hundred mRNAs. In turn, several miRNAs regulate one cognate mRNA. Thus, miRNAs function in both physiological and pathological processes, e.g., differentiation, angiogenesis, apoptosis, development of cancer, metastasis, and drug resistance. Individual miRNAs and miRNA signatures have been previously identified in monoclonal gammopathy of unknown significance (MGUS) and MM and have been described as diagnostic or prognostic biomarkers and therapeutic targets [33–48]. Global miRNA profiling studies with impact on gene expression, biological relevance, and survival have been reported, and imply a possible association with MM pathogenesis and molecular subgroups in terms of specific chromosomal aberrations or gene expression‐based high‐ risk groups. In addition, circulating and serum‐derived miRNAs have also been identified in MM
Genes 2016, 7, 84
3 of 13
patients [49–52]. In contrast, lncRNAs have been less well studied [53–56]. In the present study, we identified miRNAs and lncRNAs significantly deregulated in MM cells resistant to the three FDA‐ approved PIs, bortezomib, ixazomib, and carfilzomib. 2. Materials and Methods 2.1. Cell Lines and Reagents MM cell lines (MMCLs) were obtained from the National Cancer Institute, Bethesda, MD, USA and cultured in complete Roswell Park Memorial Institute (RPMI) media supplemented with 10% fetal calf serum and penicillin‐streptomycin. Bortezomib, carfilzomib, and ixazomib were from ActiveBiochem (Maplewood, NJ, USA). All other chemicals used were reagent grade (Sigma Chemical, St. Louis, MO, USA). 2.2. Generation of PI‐Resistant Cells RPMI8226 cells were exposed to successively increased concentrations (3 nM, 5 nM, 10 nM, 20 nM, 50 nM, and 100 nM) of bortezomib, ixazomib, or carfilzomib to generate resistant cells. Parental cells were cultured under the same algorithm in vehicle (0.05% dimethyl sulfoxide, DMSO) alone. 2.3. Cell Growth and Proliferation The effect of proteasome inhibitors on myeloma growth and proliferation was assessed by measuring XTT (Sigma) dye absorbance. First, 5 × 104 cells were plated in 96‐well plates and incubated in media that lacked phenol red. Cells were then treated with drugs at the indicated concentration (10 nM) and incubated for 72 h. XTT‐phenazine methosulfate (PMS) mixture (50 μL, 1 mg/mL XTT; 20 μM PMS) was prepared and added to plates that were then incubated for 4 h. Absorbance was then measured using a BMG Labtech FLUOstar OPTIMA plate reader (Ortenberg, Germany). 2.4. Detection of Apoptosis First, 1 × 106 myeloma cells were cultured for 24 h in a medium with or without PIs. Cells were harvested, washed, and stained with annexin V/propidium iodide (PI). Annexin V+/PI− apoptotic cells were enumerated using the Epics flow cytometer (Beckman Coulter, Indianapolis, IN, USA). The percentage of cells undergoing apoptosis was defined as the sum of early apoptosis (annexin V+) and late apoptosis (annexin V+ and PI+) cells. Annexin V fluorescein isothiocyanate (FITC) conjugate was added and the sample was analyzed using a Coulter® epics® XL‐MCL system (Beckman Coulter, Indianapolis, IN, USA). 2.5. Gene Expression Microarray Total RNA that contained both mRNA and ncRNA was isolated from patient samples using the RNeasy kit (Qiagen Inc., Germantown, MD, USA). The quality of the total RNA was confirmed using the Agilent 2100 Bioanalyzer (Agilent, Santa Clara, CA, USA) and the RNA6000 Nano assay (Agilent). For each sample, the 3′ in vitro translation (IVT) express kit (Affymetrix, Santa Clara, CA, USA) synthesized biotin‐labeled RNA target from 100 ng of the total RNA sample. The 3′ IVT kit contains hybrid primers that bind polyA (polydT) as well as random hexamer primers that bind ncRNA sequence. The kit generates cDNA to both polyA (coding) and non‐polyA (non‐coding) RNAs. A hybridization cocktail that included 10 μg of target was created for each sample. Samples were hybridized to the Genechip Primeview Human Gene Expression probe array cartridge (Affymetrix). The PrimeView Human Gene Expression Array provides comprehensive coverage of the human genome in a cartridge array format. The array is comprised of >530,000 probes covering >36,000 transcripts and variants, which represent >20,000 genes. Arrays contain probes for >20,000 total mature miRNAs, snoRNAs, ncRNAs, and pre‐miRNAs. Those probes that demonstrated a cutoff greater or less than 2‐fold from normal PCs were further analyzed.
Genes 2016, 7, 84
4 of 13
2.6. Biostatistical Analysis ncRNA profiles from drug‐naïve and drug‐resistant myeloma cells were performed and the statistical significance of the differences was determined using the Student t test with a minimal level of significance of p
